Project 1: Emerging Therapies for COVID-19: The Value of Information From More Clinical Trials
Project 2: Making Drug Approval Decisions in the face of Uncertainty: Cumulative Evidence versus Value of Information
Collaboration between: Stijntje W. Dijk, Eline Krijkamp, Natalia Kunst, Cary P. Gross, John B. Wong, Myriam G.M. Hunink
*Corresponding author, [email protected]
This code is developed with funding from the Society for Medical Decision Making COVID-19 Decision and the Gordon and Betty Moore Foundation through the COVID-19 Decision Modeling Initiative and is based on the framework from the Decision Analysis in R for Technologies in Health (DARTH) workgroup and the collaborative network for value of information (https://www.convoi-group.org/).
Project 1: Link to full manuscript in Value in Health Project 2: [Link to be added, accepted in Medical Decision Making]
The COVID-19 pandemic necessitates time-sensitive policy and implementation decisions regarding new therapies in the face of uncertainty. This study aimed to quantify consequences of approving therapies or pursuing further research: immediate approval, use only in research, approval with continued research (e.g., emergency use authorization), or reject.
Project 1 investigates the approval process of hydroxychloroquine, remdesivir, casirivimab-imdevimab, dexamethasone, baricitinib-remdesivir, tocilizumab, lopinavir-ritonavir, interferon beta-1a, and usual care.
Project 2 is a methodological review of the alternative methods to decision making in the pandemic and their performance for monoclonal antibodies (MAb) approval in hospitalized COVID-19 patients
Through this github page, we make all code available, so that our work can be adopted / built upon further. We kindly ask you to reference both manuscripts. We strongly recommend to download the entire folder and start by opening the .Rproject. The code includes all steps, from data preparation, meta-analysis, cost-effectiveness analysis, probabilistic analysis, value of information analysis to figures and sensitivity analyses.
Note that additional datasets must be downloaded from their original sources (IHME data) in order to run the full code. Instructions to do so are included within the code.